[2603.00416] MuonRec: Shifting the Optimizer Paradigm Beyond Adam in Scalable Generative Recommendation

Computer Science > Information Retrieval arXiv:2603.00416 (cs) [Submitted on 28 Feb 2026] Title:MuonRec: Shifting the Optimizer Paradigm Beyond Adam in Scalable Generative Recommendation Authors:Rong Shan, Aofan Yu, Bo Chen, Kuo Cai, Qiang Luo, Ruiming Tang, Han Li, Weiwen Liu, Weinan Zhang, Jianghao Lin View a PDF of the paper titled MuonRec: Shifting the Optimizer Paradigm Beyond Adam in Scalable Generative Recommendation, by Rong Shan and 9 other authors View PDF HTML (experimental) Abstract:Recommender systems (RecSys) are increasingly emphasizing scaling, leveraging larger architectures and more interaction data to improve personalization. Yet, despite the optimizer's pivotal role in training, modern RecSys pipelines almost universally default to Adam/AdamW, with limited scrutiny of whether these choices are truly optimal for recommendation. In this work, we revisit optimizer design for scalable recommendation and introduce MuonRec, the first framework that brings the recently proposed Muon optimizer to RecSys training. Muon performs orthogonalized momentum updates for 2D weight matrices via Newton-Schulz iteration, promoting diverse update directions and improving optimization efficiency. We develop an open-source training recipe for recommendation models and evaluate it across both traditional sequential recommenders and modern generative recommenders. Extensive experiments demonstrate that MuonRec reduces converged training steps by an average of 32.4\% while simul...